Marsh client cyber incidents more than double amid response-capacity warning
Editorial Team
••📖5 min read
Marsh reported 105 client cyber incidents in the last two weeks of October, against 48 in the final two weeks of May. The broker links the increase partly to AI, while warning that insurance panel access may not guarantee an available response team.
Marsh clients reported 105 cyber incidents in the last two weeks of October, up from 48 in the final two weeks of May.
The broker says AI accelerates exploitation of familiar weaknesses, while specialist responders face capacity constraints.
Global cyber renewal rates fell 4% in Q3; forecast growth in total premiums is a different measure.
Insurance broker and risk adviser Marsh says its clients reported 105 cyber incidents in the last two weeks of October, against 48 in the final two weeks of May. The increase comes alongside a separate warning from the broker that access to an insurer-approved response firm may not reserve a team when a business needs one.
Marsh’s company-reported client incidents in two reporting windows. The counts do not quantify AI involvement or establish an industry-wide attack rate. Source: Insurance Journal.
More incidents, without a measured AI share
The figures, disclosed by Marsh in a November 17 Reuters report, represent an increase of 57 incidents, or about 119%, between two fortnight-long reporting windows. Marsh linked the increase to the speed and scale of AI adoption and the longer time organisations take to contain attacks and respond.
That is a comparison of reported client incidents in two periods, rather than an industry-wide attack rate. The report does not provide the number of clients at risk in each window, a continuous incident series, loss amounts or a measured proportion involving AI. The increase is substantial within those reported counts; it does not establish that AI caused the entire rise.
Meredith Schnur, Marsh Risk’s US and Canada cyber practice leader, previously led the region’s cyber brokerage before her appointment to lead the regional practice from January 2024. Marsh announced that promotion alongside a global cyber practice combining insurance, risk intelligence, incident management and cybersecurity. Schnur told Reuters that attackers were showing signs of using AI-enabled and agentic tools to identify vulnerabilities. She attributed a significant number of incidents to AI without giving a numerical breakdown.
Microsoft is pitching Copilot Workspaces as a shared environment for people and AI agents. Team collaboration is already in early access, but personal Autopilot integration is still coming soon, with final agent pricing unannounced.
Editorial Team
The distinction between speed and sophistication matters. In its own November 12 response-capacity warning, Marsh said AI did not appear to be making attacks more sophisticated. Instead, it was enabling attackers to exploit unpatched systems and weak identity controls at greater speed and scale, and to maintain access longer. The broker urged organisations to review patching, vulnerability management and identity access controls.
The bottleneck is access to qualified responders
Marsh describes a sustained rise across third-party threat activity, ransomware, privacy incidents and vendor-driven events, rather than a single catastrophe. It says response firms cannot hire and train qualified staff quickly enough, while law firms and extortion negotiators face similar pressure.
The skills mix compounds that constraint. Marsh says cloud infrastructure and web application incidents have become a significant area of impact, while some response teams lack deep cloud expertise. The pool of suitable responders can therefore be smaller than total staffing suggests.
An insurer’s panel can supply vetted providers, negotiated rates and a simpler engagement process. Marsh warns, however, that panel membership alone does not necessarily allocate a team to a policyholder. Paid retainer clients may receive priority under contractual response-time commitments; other clients may receive help only as capacity allows, or have their incident referred to another firm. The broker also says some response windows are lengthening, but supplies no market-wide measure of the delay.
Retainers can also include advance planning, incident exercises and access to a team familiar with the client’s systems. Those services, panel access and a contractual response time are distinct benefits; buyers need to establish which their arrangements actually provide.
This warning also comes from a business with a commercial stake in the response market. Marsh’s advisory offers to review insurance arrangements, evaluate retainers and connect clients with cybersecurity vendors. Its account helps explain what buyers should examine, but the published warning does not quantify the industry’s responder shortfall or establish that every organisation needs a retainer.
GlobalData’s company-reported forecasts concern aggregate cyber insurance premium volume, not an individual buyer’s renewal price. Source: Insurance Journal.
Premium volume is not a policy price forecast
GlobalData forecasts global cyber insurance premiums of $16.6 billion in 2026, up 6.9%, followed by 10% growth to $18.3 billion in 2027, according to the report on Marsh’s incident figures.
Those are forecasts for aggregate premium volume. They do not establish a 10% increase in a particular buyer’s renewal price, nor do they isolate price changes from changes in coverage purchased.
Marsh’s own third-quarter 2026 market index, which measures commercial insurance rate changes at renewal, shows global cyber rates fell 4%, matching the previous quarter. US cyber rates were flat, while other regions continued to see reductions. These are average rate movements reflecting Marsh’s client portfolio mix, rather than a quote for any individual buyer.
The company’s accompanying October 21 announcement called it the thirteenth consecutive quarter of global cyber rate declines. Cyber insurance capacity continued to expand through new entrants, and buyers could use the added capacity to broaden coverage and lower the losses they retain themselves.
The index also described steady cyber claim frequency and rising severity in Q3. That earlier market account and the later comparison of two client-incident windows measure different periods and outcomes: reported incidents are not interchangeable with insurance claims. Neither supplies a continuous series establishing how industry-wide attack frequency changed between May and October.
Insurance capital and specialist response labour are different resources. The evidence allows falling renewal rates and tightening responder availability to coexist; a cheaper policy does not by itself answer how quickly a qualified team will engage.
What buyers need to establish before the next incident
Marsh recommends checking whether existing arrangements promise a contractual response time or merely panel access, whether a preferred provider has insurer approval, and who takes over if the first-choice firm cannot respond. It also recommends testing vendor availability during incident exercises and assessing retainers against the organisation’s risk and budget.
For the broader AI claim, the missing evidence is a consistent incident series with client exposure, severity and an explicit method for identifying AI involvement. Until those details are available, Marsh’s figures support a sharp increase in reported client incidents and a warning about response capacity. They leave the size of AI’s contribution unresolved.
Google is rolling out file references and sharing for Gemini skills as adult personal accounts lose the ability to use Gems. Automatic conversion preserves supported material, but file limits and missing tools can interrupt existing workflows.
Muddy Waters disclosed a short position in Nvidia while questioning a possible connection involving senior adviser Ajay K. Puri. Its own account leaves the key identity question unanswered, separating its new claims from an existing server-diversion prosecution.
Nvidia reported $115.8 billion in quarterly revenue in its first earnings release since co-founder Jensen Huang resigned. Strong demand gives interim CEO Colette Kress a financial cushion, while a campaign for Huang’s return collides with unresolved investigations and the board’s reported opposition.
OpenAI launched GPT-6.2 Astra and Sol for professional work, cutting Astra’s standard input and output token prices by 20%. The models offer cheaper paths through selected tasks, but research evaluations, reporting failures and unfinished independent safeguard testing complicate the case for broader autonomy.
Microsoft’s first live speech-to-speech model offers OpenAI API compatibility and planned enterprise integrations. Its published evaluations show a low-cost option, with several rivals completing more agentic tasks.
Canberra-based Trellis Data is offering its 220-billion-parameter Agentaus 1 model to business and government. Its pitch goes beyond keeping data in Australia, but performance, deployment costs and a planned capital raising remain key tests.
Perplexity is adding direct collaboration between frontier models to Computer, letting a lead agent coordinate up to 16 supporting agents. Company-reported benchmarks suggest mixed teams can outperform repeated copies of one model, but leave customer costs and production reliability unresolved.
Meta’s Muse personal AI agent has launched in Australia, with some local shopping and booking connections to follow. Its browser and payment tools broaden its reach, while the company’s technical disclosures qualify its privacy promises.
Microsoft and Salesforce are extending their sales software alliance with Copilotforce, combining CRM workflows with Microsoft 365 context under existing permissions. The Copilot plugin enters public preview, while Teams remains in pilot; earlier Microsoft tools and a Claude beta raise the bar for proving its value.
Microsoft expands MAI Image 2.7 from an editing preview into image generation and API access, with cheaper Flash pricing and a Canvas workspace. Benchmark claims depend on the workload, while the MAI Labs interface carries restrictions on commercial use.
OpenAI is widening applications for its legal AI configuration after improving research scores and expanding its search index. Access still requires approval, and benchmark results show substantial limits alongside the gains.
Microsoft has made Word, Excel and PowerPoint creation generally available inside Copilot, combining editable files with workplace permissions and collaboration. The broader shift toward autonomous work remains staggered across previews, while its claimed cost advantage rests on company-funded testing.
Knit Health’s NEA-led Series A funds deployment of Med-BEAM for care navigation and hospital operations. Its early company benchmark beats three Claude models at predicting clinical actions, while better patient outcomes remain unproven by that test.
Science Corporation’s Series D follows its European retinal-implant launch. Clinical results support improved central vision, but U.S. approval, eligible patients and commercial terms remain consequential questions for the startup, reportedly valued at $3.5 billion.
Google will let Gemini users save conversations while opting out of generative AI training on future chats. The global rollout also extends storage of history-off and temporary chats from 72 hours to around two months; human review and longer retention can still apply.
Google is removing the subscription requirement for Gemini’s task mode in six countries, while adding U.S. payment, calling and messaging features. The expansion broadens access to delegated work, but usage limits, permissions and website compatibility still govern what it can finish.
Eligible Pixel owners can join a Play Store beta to generate home-screen widgets from descriptions, but device, language and regional restrictions still apply.
Microsoft’s new in-house coding models compete with outside suppliers inside GitHub Copilot. Company tests show lower task costs at promotional prices, while its continuing financial ties to OpenAI complicate the rivalry.
Valkryn disclosed a $15 million seed round and early military testing contracts. Its evaluator-focused Battle Lab faces existing commercial tools, public testing capabilities and funding constraints that software alone cannot resolve.